On Sat, 2005-04-16 at 16:32 +0200, Jakob Bohm wrote:
> On Mon, Apr 11, 2005 at 07:50:36AM -0400, Anthony DeRobertis wrote:
> > Ingo Ruhnke wrote:
> >
> > >Sound free to me, since not the output of the library is required to
> > >confirm to it, but the interface which generates the input for the
> > >library. So to me it looks basically just like something like GPLs 2c
> > >section applied to the web.
> >
> > Not really. GPL 2(c) tells me what changes I may make to the software in
> > question. This tells me what changes I may/must make to other software.
> >
>
> GPL 2(c) is the obnoxious banner clause that says that if you
> take a random piece of non-interactive GPL code and incorporate
> it in an interactive program, then that program must include a
> startup banner telling the users that this is Free Software
> under the GPL.
Right, and this is relevant if and only if:
1) the original GPL work printed such an announcement; and
2) the relevant work is a derivative work of both GPL'd code and code
that prints such a notice.
> Similarly, cl-typesetting #5, says that if you incorporate
> cl-typesetting in a larger program then that program (not its
> output) must display a cl-typesetting banner in its primary
> startup screen.
Ah, but this is not a derivative work of cl-typesetting. cl-typesetting
(I am assuming) merely processes some marked-up data.
> GPL 2(c) presumes a command-like program similar to
> gdb/bash/emacs in its example text, cl-typesetting#5 presumes a
> HTML-based user interface such as a cgi/php/jsp frontend.
Yes, but the HTML-based interface is a derivative work not of
cl-typesetting, but of the input. The GPL'd program we are discussing is
a derivative work of only the works which make up the executable. cl-t
#5 would contaminate other software, specifically the input to the
typesetter.
> So one must look very carefully to determine what places GPL
> 2(c) just within the DFSG (other than DFSG#10), and what causes
> cl-typesetting #5 to be within or outside the DFSG.
You are comparing apples to oranges. The GPL'd program is a derivative
work under the GPL; the postprocessed text is not a derivative work of
cl-typesetting, unless cl-typesetting copies significant portions of its
own code into the output, like bison does[0].
As an analogy, I would like to point out that I am writing free
documentation (how it is licensed is not really relevant) with troff
markup. Does my documentation then become a derivative work of groff?
I hope not. Otherwise, once sarge releases, all the BSD manpages with
advertising clauses would become undistributable.
That said, I would like to point out that GPL 2(c) is not a favorite of
many people on this list. It is my belief, however, that any
interpretation of the DFSG (ignoring section 10) which would make the
licenses in DFSG 10 non-free is an incorrect one.
[0] Bison now has an exception for the output.
--
($_,$a)=split/\t/,join'',map{unpack'u',$_}<DATA>;eval$a;print;__DATA__
M961H<F$@8FAM;"!U<F%O<G-U(#QU<F%O<G-U0&=D:75M<&UC8VUL=G)U;6LN
M<FUL+F=Y/@H)>2QA8F-D969G:&EJ:VQM;F]P<7)S='5V=WAY>BQN=V]R8FMC
5:75Q96AT9V1Y>F%L=G-P;6IX9BP)
Attachment:
signature.asc
Description: This is a digitally signed message part